Low Energy Sketching Engines on Many-Core Platform for Big Data Acceleration

Abstract

Almost 90% of the data available today was created within the last couple of years, thus Big Data set processing is of utmost importance. Many solutions have been investigated to increase processing speed and memory capacity, however I/O bottleneck is still a critical issue. To tackle this issue we adopt Sketching technique to reduce data communications. Reconstruction of the sketched matrix is performed using Orthogonal Matching Pursuit (OMP). Additionally we propose Gradient Descent OMP (GD-OMP) algorithm to reduce hardware complexity. Big data processing at real-time imposes rigid constraints on sketching kernel, hence to further reduce hardware overhead both algorithms are implemented on a low power domain specific many-core platform called Power Efficient Nano Clusters (PENC). GD-OMP algorithm is evaluated for image reconstruction accuracy and the PENC many-core architecture. Implementation results show that for large matrix sizes GD-OMP algorithm is 1.3× faster and consumes 1.4× less energy than OMP algorithm implementations. Compared to GPU and Quad-Core CPU implementations the PENC many-core reconstructs 5.4× and 9.8× faster respectively for large signal sizes with higher sparsity.

Programmable Big Data Processing Framework to Reduce On-Chip Communications and Computations Towards Reducing Energy of the Processing

“…In this particular style, every 3 primaries along with a moment are actually arranged right into a cluster which can easily execute intra-communication straight with a bus and also inter-communication via a transmitting style. The many-core is actually entirely executed in Verilog and also put and also directed in 65nm, 1V CMOS Technology [4] Figure 4B reveals the style of a bus-based cluster of PENC several-primary platform. Our many-core progression setting features an architecture simulation filled in Espresso.…”

“…The task element is actually after that obtained and also is actually made use of due to the Tempo Conflict device for precise energy calculation. The PENC many-core design is actually preferably matched for many biomedical requests as it attends to the features as well as difficulties fundamental along with these uses [4], [Thirteen] [14].…”

“…In this paper, we suggest a sketching-based big data processing structure that carries out all signal processing in your area at the sensor. We incorporate a low-overhead mapping out structure with domain name details programmable PENC many-core accelerator to reduce calculation power [4] Whereas Mapping out strategy at the front end lowers variety of data transfers, calculations, and also storage. The laying out algorithm decreases data by getting a linear mix of the data prior to processing.…”